Executive Summary: Below 10 Lives Cannot Be Insured Through Any Group Mechanism
TOS.09 — The Other Side#
Below a threshold that actuarial science places somewhere in the range of 25 to 50 lives for even minimal statistical credibility, group coverage is not a product refinement problem. It is an actuarial impossibility disguised as a product problem. The products sold to groups of 2 to 10 employees are not group insurance in any meaningful sense. They are annual financial wagers dressed in insurance language, with stop loss carriers as the house and employers as players who do not know the odds are incalculable at the group size they represent.
Insurance pricing depends on credibility: the degree to which a group’s own claims experience can predict future claims. For groups of 25 to 50 members, actuaries treat claims experience as having partial credibility, blended with population-level manual rates. Below 25 members, the experience weighting drops sharply. For groups of 10 or fewer, industry practice weights almost entirely on manual rates because the group’s own experience is actuarially noise. The NAIC’s white paper on stop loss insurance and self-funding states explicitly: “In the case of a very small group, a credible estimate of expected losses may not be realistic. In these circumstances, an actuary may be unable to determine, with a reasonable degree of actuarial certainty, the expected claims of the small employer.” The products are built anyway.
At five lives, one member is 20 percent of the risk pool. A diagnosis of multiple sclerosis and treatment with Tysabri at approximately $76,000 per year, or a premature birth requiring neonatal intensive care at costs commonly exceeding $100,000 per event, defines the plan year around a single event no actuary could price with statistical confidence. What the carrier actually prices when underwriting a group of 5 to 10 employees is demographic risk scoring, not group claims experience analysis. Most carriers apply a minimum percent-to-manual floor of 50 to 70 percent regardless of how favorable the demographics appear. The employer is paying primarily for the carrier’s uncertainty margin, not for an accurate assessment of their specific risk.
EBRI data documents the market outcome: the percentage of employers with fewer than 10 employees offering health benefits declined from 34.2 percent in 1996 to 22.5 percent in 2023, nearly 12 percentage points over 27 years. The market is already solving the problem in the direction the actuarial evidence points. Micro-employers are not finding better group products; they are leaving group coverage behind.
Two existing mechanisms serve this population more honestly. The QSEHRA, created by the 21st Century Cures Act in December 2016, allows employers with fewer than 50 full-time equivalent employees to reimburse employees for individual market premiums on a tax-free basis, with 2025 contribution limits of $6,350 for single coverage and $12,800 for family coverage. The ICHRA offers the same mechanism with class variation and without contribution caps. Neither involves a group plan, claims adjudication, stop loss, or an actuarial credibility problem. The employer’s exposure is a defined contribution. The honest policy conversation shifts from how to extend group coverage downward to what the right mechanism is for this population, and what the transition costs the employees currently served by micro-employer group plans.